Don’t follow me too closely

tdq6bvpwadnmy43odarp

It occurred to me that if my super-picker was very successful, then my @000 predictions would start to get fed into the super-picker, causing a preaching-to-the-converted feedback loop: I would lose 1/24th of my unbiased information, assuming I am picking top 24 forecasters.

If in addition another forecaster copied all of my forecasts, then I would lose 2/24ths or 8.3% of my information.

So in super-picking I will need, upon eventual wild success, to filter out forecasts from  @000 and any forecaster whose deviation from @000 forecasts is below some tolerance.

Separately from that, please note that I am once again winging it.  If I think that Ted Cruz doesn’t have a chance in hell of being nominated for President, then my forecast goes to 0, even though the consensus of supers is 34%.

Note that I bet against Bibi last year and in favor of Jonathan Goodluck’s candidacy, so it would most likely not pay to follow me too closely here, unless you are a fan of PowerBall.

Advertisements

2 thoughts on “Don’t follow me too closely

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s